1. Technical Field
The present invention relates to an object detection apparatus for detecting a target object from an input image with use of an image recognition dictionary.
2. Related Art
A known object detection apparatus for detecting a target object that is an object to be detected, such as a person or a vehicle present forward or rearward of a subject vehicle, applies a predetermined image recognition algorithm using an image recognition dictionary describing reference data for the target object to an image captured by and received from a camera or the like. For example, an object detection apparatus as disclosed in Japanese Patent Application Laid-Open Publication No. 2005-259031 recognizes the presence of a person by setting a reference pattern for recognizing the presence of a person according to external environment information, such as weather or temperature information.
The image recognition dictionary used in the image recognition of the target object is predefined for a specific assumed environmental condition. However, if an environmental condition when the image was captured is far different from the environmental condition assumed when the image recognition dictionary was predefined, robustness of the image recognition results obtained by using the image recognition dictionary will be reduced and thus performance of detecting the target object will be lowered.
In the disclosed apparatus, setting the reference pattern in response to changes in the external environment information about the weather or temperature or the like can enhance the detection performance. In such an object detection apparatus mounted in a vehicle, however, changes in environmental condition caused not only by the changes in the external environment information, but also by changes in operating state of the subject vehicle, may affect the detection performance.
The visibility of the target object is variable. For example, the whole body of a pedestrian is visibly bright in the daytime hours when headlights of the subject vehicle are off while a lower part of the body of the pedestrian is visibly bright and an upper part of the body of the pedestrian is too dark to be seen in the nighttime hours when the headlights of the subject vehicle are on. Hence stable performance of detecting the target object may not be achieved when the image recognition that uses the image recognition dictionary and the image recognition technique predefined for the specific assumed environmental condition is applied to the images captured in various changing environmental conditions around the target object caused by the different operating states of the subject vehicle including active and inactive states of the headlights of the subject vehicle.
In consideration of the foregoing, it would therefore be desirable to have an object detection apparatus mounted in a vehicle, capable of reliably detecting a target object even in various changing environmental conditions caused by changes in operating states of the vehicle.